This renewal application seeks support to complete the 10-year follow-up of the Suffolk County Mental Health Project (SCMHP). This first-admission cohort was recruited between September 1989 and December 1995 primarily from the public, community, and private hospitals of Suffolk County, Long Island (population 1.3 million), diagnosed at baseline, 6- and 24-month follow-up using a longitudinal consensus procedure, and reassessed on multiple occasions over a 4-year period. The three largest DSM-IV diagnostic groups were schizophrenia, bipolar disorder with psychotic features, and major depressive disorder with psychotic features. In reevaluating the cohort at the 10-year point, the study extends our understanding of the stability of diagnosis, course of illness, quality of life, and treatment patterns by wedding clinical and epidemiological methods in a single, longitudinal study. [unreadable] [unreadable] A total of 581 members of the SCMHP cohort are being targeted for the 10-year follow-up. Follow-up assessments are being conducted by master's level mental health professionals trained on the SCID and other instruments used for the study. The respondent interview is conducted in two sessions. Session I focuses on interval episodes of illness, quality of life, trauma exposure and PTSD, social support and risk behavior. Session II focuses on finances, treatment, neuropsychological functioning, co morbid disorders, family history, physical health, and personality. A single interview is conducted with a significant other and addresses respondent's current symptoms, family history of mental illness, psychosocial functioning, family burden, perceptions of stigma, and perception of the adequacy of mental health services received by the respondent and by the significant other. Standard clinical ratings and a narrative are, completed by he interviewer after all inpatient and outpatient records are obtained. The diagnostic information is reviewed by, the project psychiatrist, and conferences are held to confirm the diagnosis, formulate consensus ratings of illness course, trauma and PTSD. As before, most interviews are conducted in respondents' homes. A number of substantive and methodological papers have been published from the study in collaboration with researchers from several institutions. As we enter the new era of molecular epidemiology, it is our hope that our sample, which is the only current epidemiological sample in North America, will prove to be a rich resource on which to build future studies.